Update README.md

README.md

@@ -67,29 +67,45 @@ pip install -e ".[vllm]"
 
 Quickstart with Python:
 ```python
-from flow_judge.models.model_factory import ModelFactory
-from flow_judge.flow_judge import EvalInput, FlowJudge
-from flow_judge.metrics import RESPONSE_CORRECTNESS_BINARY
+from flow_judge import Vllm, Llamafile, Hf, EvalInput, FlowJudge
+from flow_judge.metrics import RESPONSE_FAITHFULNESS_5POINT
 from IPython.display import Markdown, display
 
-# Create a model using ModelFactory
-model = ModelFactory.create_model("Flow-Judge-v0.1-AWQ")
+# If you are running on an Ampere GPU or newer, create a model using VLLM
+model = Vllm(quantization=True)
+
+# If you have other applications open taking up VRAM, you can use less VRAM by setting gpu_memory_utilization to a lower value.
+# model = Vllm(gpu_memory_utilization=0.70)
+
+# Or create a model using Llamafile if not running an Nvidia GPU & running a Silicon MacOS for example
+# model = Llamafile()
 
 # Initialize the judge
-judge = FlowJudge(
-    metric=RESPONSE_CORRECTNESS_BINARY,
+faithfulness_judge = FlowJudge(
+    metric=RESPONSE_FAITHFULNESS_5POINT,
     model=model
 )
 
-# Prepare evaluation input
+# Sample to evaluate
+query = ...
+context = ...
+response = ...
+
+# Create an EvalInput
+# We want to evaluate the response to the customer issue based on the context and the user instructions
 eval_input = EvalInput(
-    inputs=[{"question": "What is the capital of France?"}],
-    output="The capital of France is Paris."
+    inputs=[
+        {"query": query},
+        {"context": context},
+    ],
+    output={"response": response},
 )
 
-# Perform evaluation
-result = judge.evaluate(eval_input)
-print(result)
+# Run the evaluation
+result = faithfulness_judge.evaluate(eval_input, save_results=False)
+
+# Display the result
+display(Markdown(f"__Feedback:__\n{result.feedback}\n\n__Score:__\n{result.score}"))
 ```
 
 Discover more at our repository [https://github.com/flowaicom/flow-judge](https://github.com/flowaicom/flow-judge)